COOL grocery shopping: Swedish consumers' perception of country of origin information and labels in an online shopping environment

DA234X Degree Project in Computer Science and Engineering, specializing in Media
 Management
 Second Cycle 30.0 credits

 COOL grocery shopping: Swedish
 consumers’ perception of country of
 origin information and labels in an
 online shopping environment
 SANDRA SJÖÖ

Stockholm, Sweden 2022

TRITA – XXX-XXX 20XX:XX

www.kth.se

COOL grocery shopping: Swedish consumers’ perception of country of
origin information and labels in an online shopping environment

SANDRA SJÖÖ, KTH Royal Institute of Technology, School of Electrical Engineering and Computer Science,
Sweden

 Though grocery shopping is largely done by habit, consumers are provided with continuous information aiming to
inform and persuade them to make active choices. Online grocery shopping offers further ways for consumers to find,
evaluate and compare products with just a few clicks. Country of origin labelling (COOL), providing consumers with
further information, has gained popularity over recent years. What influences consumers to actively choose COOL products
has been researched, but specifically Swedish consumers’ evaluation of COOL online has remained unexplored.
 The aim of the current study is to contribute to the understanding of consumers’ evaluation of COOL in the online
environment, particularly for Swedish consumers. A survey was distributed and the answers analysed in order to evaluate
Swedish consumers’ perceptions of COOL in the online environment.
 The results suggest that Swedish consumers value COOL highly, since it aligns with their values. Those who have a
higher interest in COOL pay less attention to price, and they are willing to pay more for Swedish products. Additionally,
more ways to find and filter products online based on country of origin, is desired amongst consumers who value COOL.

SAMMANFATTNING
Trots att mathandel ofta görs på ren vana, så förses konsumenter med massor av information med syfte att informera
och övertyga de att göra aktiva val. Mathandel online erbjuder ytterligare sätt för konsumenter att hitta, utvärdera och
jämföra produkter med endast ett par klick. Ursprungsmärkning, som förser konsumenter med än mer information, har
vuxit i popularitet de senaste åren. Det har forskats kring vad som influerar en konsument att aktivt välja ursprungsmärkta
produkter, men svenska konsumenters värdering av ursprungsmärkning i onlinehandeln har förblivit outforskat.
 Syftet med denna studie är att bidra till förståelsen för konsumenters värdering av ursprungsmärkning i onlinemiljö,
specifikt med hänsyn till svenska konsumenter. En enkät användes för att utvärdera och analysera svenska konsumenters
uppfattning av ursprungsmärkningar i onlinemiljö.
 Resultatet antyder att svenska konsumenter värderar ursprungsmärkta produkter högt, då de efterlever deras värderingar.
De som har ett större intresse för ursprungsmärkningar lägger mindre vikt vid pris och är villiga att betala mer för svenska
produkter. Dessutom efterfrågas fler tillvägagångssätt för att hitta och filtrera produkter online baserat på ursprung bland
konsumenter som värderar ursprungsmärkningar.
Key Words and Phrases: country of origin (COO), country of origin labelling (COOL), consumer values, online grocery shopping

Nyckelord: ursprungsinformation, ursprungsmärkning, konsumentvärden, mathandel, online, svenskmärkning
Author’s address: Sandra Sjöö, sasjoo@kth.se, KTH Royal Institute of Technology, School of Electrical Engineering and Computer Science, Stockholm,
Sweden, SE 100 44.

Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or
distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for third-party components of
this work must be honored. For all other uses, contact the author.
Document date: May 24, 2022
Swedish title: COOL livsmedelshandel: Svenska konsumenters uppfattning av ursprungsinformation och märkningar i onlinemiljö
© 2022 Copyright held by the author.

 1

Sandra Sjöö

1 INTRODUCTION
Grocery shopping is a elemental part of our everyday lives. As we select our groceries, we rely on a range of different cues,
such as price, appearance, country of origin and various labels [10, 42]. These cues have been researched for decades in
the context of physical grocery shopping. However, grocery shopping is increasingly taking place online. In 2021, online
grocery shopping represented 87 percent of the 3.7 billion SEK growth for grocery shopping in Sweden [28], and online
grocery shopping is a ever growing phenomenon [28, 60].
 As products are now often viewed through screens, rather than physically in front of the consumers, the grocery field,
and subsequently the consumers’ perception of the cues, is changing. Online, consumers rely primarily on the visual
information that is presented to them, such as images and icons, and secondly on the textual information [14]. As products
are typically displayed with equally sized images on pages filled with various products [17], and a lot of information is
competing for the consumers’ attention, important information can also be missed. Additionally, consumers tend to not
regard information that is a click away at all [14], and their decisions are based on what they initially perceive. If cues are
not clear, arguably, this leads to uninformed decisions.
 Country of origin as a cue plays a significant role for the consumer when evaluating unfamiliar products [45]. For many
types of food it is mandated by laws and regulations for the producers to inform consumers about the country of origin
[30]. Country of origin labels (COOL) – distinct markings on products that clearly communicate the country of origin –
are increasingly used as a way to inform consumers and make products more desirable. Indeed, it has been shown that
consumers perceive that food with such labels tastes better [32]. Additionally, COOL aims to provide consumers with
further information about products’ transportation distances and production methods, which can become more evident
when the consumer knows where the product originates [49].
 Previous research states that COOL matters to consumers for various reasons, and that their values, such as caring about
sustainability or supporting the national economy, play a specific role with regards to COOL [32]. Still, some research
shows that while the number of various labels is increasing, consumers’ understanding of them is decreasing [49]. These
trends, combined with the various ways online stores display COOL [47], mean that there is great variation in consumers’
perception of labels.
 With online grocery shopping increasing it is interesting – from a research point of view, but also a producer, label,
grocery store and society point of view – to investigate the correlation between consumers’ interest, perception of and
attitudes toward COOL in the online environment, as well as who chooses COOL products and why. This dissertation
aims to investigate the following research question: How do Swedish consumers perceive and value COOL products and
information in online grocery shopping environments?

2 LITERATURE REVIEW
In this section, the literature that has been considered for the current study is outlined. Topics of online grocery shopping,
information structure and processing, country of origin labels, cues, values and demographics were deemed relevant.

2.1 Grocery shopping off- and online
Just like physical grocery stores, online stores compete for consumers in four main ways: pricing, product assortment,
convenience and service [11]. Several studies on online grocery shopping show that the major reasons why people begin
shopping online include saving time, caring for a sick relative, moving, convenience and avoiding long lines in physical
stores [21, 31, 38, 63]. Anesbury et al. argue that those who choose to shop for groceries online tend to “be younger and
 2

COOL grocery shopping: Swedish consumers’ perception of country of origin information and labels in an online shopping environment

have higher incomes” [12, p.263]. In Sweden, 10 percent of the population shops groceries online every other week or
more often, however, over 60 percent of Swedes have never purchased groceries online [36]. Still, online grocery shopping
is growing faster than grocery shopping in physical stores [28, 60].
 The majority of online grocery stores today are extensions of already established physical stores [11]. Currently, in
Sweden, the biggest online grocery retailers are ICA (owned by ICA Group [4]) with over 50 percent of the market share,
Mathem, Mat.se, City Gross, Willys and Hemköp (owned by Axfood and Bergendahls [1–3]) with about 24 percent, and
Coop (owned by Coop [72]) with just over 18 percent [37, 53].
 Physical grocery stores, where consumers navigate aisles of products, are planned and designed to help consumers find
and purchase products, and to inspire and encourage additional purchases, using methods such as visual merchandising
and product displays [22, 23]. While online stores are largely built based on the same model [14], there are, however,
major differences between the two. In a physical store, the consumer can pick up the products and use additional senses,
such as smell and touch. Moreover, there are physical staff available to help the consumer locate and navigate amongst
options. Within an online store, the products are represented by images and text on a screen, and the consumer relies
completely on their vision and the access to the information [47]. However, with online shopping, the consumer can pause
their shopping at any time and check their pantry if they need to buy certain products, whilst shopping. Online stores lack
staff to assist the consumer if they struggle to find what they are looking for.
 Generally, consumers who shop online tend to behave as they do when shopping in physical stores when they look for
products, with the majority choosing to browse through the menus of product categories [14, 17, 54]. Consumers behave
in similar ways possibly because shopping for groceries is, in fact, a low-involvement activity, and when browsing through
the menus the consumer relies on recognition of products, rather than recall [14]. This method also offers consumers the
possibility to evaluate alternatives in the chosen product categories [14]. Benn et al. found that the consumers in their
study used the search function only when they failed at finding their desired products through the menus [14], which could
be equated to asking an employee for help.
 It is common for grocery stores to offer loyalty programmes for their customers, to collect points and bonuses when
shopping, both online and in physical stores. Online shopping allows extensive individual information to be gathered,
enabling completely tailored product selections, recommendations and pages for each consumer, based on their previous
purchases and favourite-marked items [59, 61, 77]. Olzenak et al. [58] means that this creates opportunities for online
stores to further develop and personalise the filtering functions within the menus, based on the consumer’s previous choices
and preferences. Berg et al. [17] also argue that the online environment enables further opportunities for consumers to
quickly search and filter for certain attributes, such as country of origin, and that this type of product metadata, and how
products are presented, can have a big impact on which products the consumer finds and chooses.
 When shopping for groceries online, consumers make their product evaluations firstly based on visual information (e.g.,
images) and secondly through the first available textual information (e.g., descriptions), and any information that is an
additional click away is rarely viewed at all [14, 17, 58]. As some senses, such as touch and smell, cannot be used in
the product evaluation, many consumers are hesitant towards buying fresh produce when shopping groceries online, in
comparison to a physical store [31, 61].
 The information available online varies greatly compared to physical stores, which can lead to consumers not finding
the desired information to base their decisions upon [58]. In their investigation of 555 product searches in seven different
grocery retailers in Canada, Lee et al. [47] found that the presentation of products and available information, particularly
that of different labels, varied significantly. Research on product imagery online shows that when displaying details and

 3

Sandra Sjöö

labels on product packages, consumers who are not familiar with a brand regard the labels positively, both in terms of
assessing a product’s size and physical attributes, as well as their attitudes toward the brand and product quality [15, 16].
 Additionally, when presenting products online, the images used are often the same size, despite differences in actual
product sizes, which can lead to difficulties and frustrations for the consumers as they are browsing the products [16, 17].
Online, the stores have great control over product presentation and imagery shown, while the producers and product
brands have very little influence. Thus, the online stores have complete control over how any potential labels are, or are
not, displayed to the consumers.

2.2 Country of origin information and labels in Sweden
Within the current study, ‘labels’ refers to a distinct marking on a product that clarifies specific information about the
product, such as country of origin or method of production, i.e., not the general product information on the packaging.
According to the Swedish Food Agency, origin means “produce completely produced within one country or territory is to
be considered originating from that country or territory” [30]. Thus, country of origin labelling (COOL) refers to a label
which communicates the country of origin of the product to the consumer.
 Labelling of foods and products with their origin can be done on a regional or country level, or for a larger geographic
area, such as the EU. In the Swedish and EU contexts, consumers prefer country of origin over both region of origin and
the EU/non-EU alternative [25, 70]. Legislation within Sweden enforces the availability of origin information on country
level for meat, fish, honey, fruits and vegetables [30]. For other kinds of food, it is voluntary to include country of origin,
unless consumers risk being misled [30].
 As an addition to the mandatory information, producers can label their product with independent labels in order to
enhance the products’ aspects, such as country of origin [44]. The "Från Sverige" ("From Sweden") label is a voluntary
Country of Origin Label for Swedish-produced raw produce, foods and plants [7]. The label is owned and managed by
Svenskmärkning AB, which was co-founded by three organisations: Lantbrukarnas Riksförbund, a non-political interest
and business organisation; Livsmedelsföretagen, the industry and employer organisation for food companies in Sweden;
and Svensk Dagligvaruhandel, the Swedish trade association for the grocery trade. The three owners run Svenskmärkning
AB, and thus “Från Sverige”, with equal shares, but without profit: all profits made from licences go back to the companies
who use the label, for marketing and communication purposes.
 Svenskmärkning AB was founded to support the Swedish food industry by providing consumers with information about
food and other products, such as flowers, from Sweden [5]. In order to qualify for the label, a product must be grown or
born and raised, refined, packaged and controlled in Sweden, depending on the type of product. This is ensured by an
independent control agency [7], which performs random revisions of the producers each year [6]. The aim of the label is
to increase awareness of products that fulfil the criteria, and thereby encourage consumers to choose these products [7].

2.3 Legitimacy and trust in labels
Research shows that consumers’ general understanding of the growing array of different kinds of labels – country of origin
labels, eco-labels, nutrition labels, regulation labels, etcetera – is low and decreasing at the same time as the number of
different kinds of labels – and other ways to communicate the country of origin and product quality such as imagery, flags,
symbols or even statements – are increasing [49, 75]. Despite the growing use of different labels, there is a huge lack of
quality control, with some being owned by the brands themselves and thus not controlled by an independent party [35, 75].
Amongst EU consumers, just above 12 percent believe that labelling provides an additional reassurance of quality, and
 4

COOL grocery shopping: Swedish consumers’ perception of country of origin information and labels in an online shopping environment

about 10 percent consider the safety of the food guaranteed with an origin label [25]. However, it is shown that in countries
with high trust in institutions and governments, such as Sweden, consumers have a higher trust in such labels [73, 74].
 Amongst the Swedish population, consumers generally believe country of origin is an important quality cue for food
products, and women value it to some higher extent than men [32, 36, 42]. Producers are generally positive towards origin
labels [18, 35], but also highlight that control is important to ensure legitimacy [35]. The producers’ perception of origin
labelling differs somewhat to that of the consumers, as the producers believe that consumers value the local connection
above a specific label [35]. Fredriksson and Norberg [35] discuss that the producers believe that prioritising one known
label (typically one of the eco-labels) provides enough information to the consumers, who may otherwise get confused or
overwhelmed, and that for the smaller producers it can get too costly to maintain several certifications.

2.4 COOL and its effects
The aim of country of origin labels is to provide consumers with additional information, for instance about transportation
distances and production methods, which can become more evident when the consumer knows where the product originates
[49]. Certain attributes can be valued favourably based on the country of origin, for example the quality of the products
[51], while others are regarded less favourably, such as animal welfare conditions [42, 45]. In that sense, COOL provides
more information in its own right, as well as adds a layer of information on top of other attributes [49, 75] and therefore
enables consumers to make more informed decisions.
 Research on COOL often also discusses organic and ecological food and labels, here referred to as the Swedish
denomination ecological and/or eco-labels. Research on COOL and eco-labels often treats them as interchangeable
and complementary. They are in many ways similar, as a means to provide the consumer with more information about
production methods, but the two are distinct. Similar to COOL, regulations vary in different countries for which produce
may be labelled as ecological. Generally, organic or ecological food, here referred to as the Swedish ecological food, can
be defined as, but not limited to, fruit and vegetables grown without unnatural pesticides, meat from animals who have
been fed with ecological feed and allowed to live as naturally as possible with minimal use of antibiotics, and produced
completely free from genetically modified organisms [9]. Hence, eco-labels are labels that aim to guarantee that the current
regulations within the country of origin are fulfilled by the producer.
 The two labels are often connected since COOL is notably affected by, and, in turn, affects eco-labels. Research into
the effects of COOL and eco-labels in relation to one another show that country of origin information of a product can
influence consumers’ perceptions of the eco-label, since the country may have strict regulations regarding what can be
labelled ecological [74, 75]. Thus, COOL both complicates and supplements eco-labels, and is at times even preferred over
it [41]. Thøgersen et al. [75] argue that COOL is assigned less value for a single product when an eco-label is available,
but when comparing two ecological products, country of origin plays a significant role in consumers’ decisions.

2.5 Cues and product attributes
When evaluating products, consumers make their decisions based on different product attributes, or cues, which are of
different value to different consumers depending on, for example, socioeconomic factors and previous experiences [42].
Cues are either intrinsic or extrinsic “informational stimuli available to the consumer” [10, p.104]. When intrinsic cues,
such as taste, are difficult to assess, consumers rely on extrinsic cues, such as price or country of origin, instead [10].
 Research shows that while country of origin matters as an extrinsic cue [42], it tends to have less value to the consumer
when other extrinsic cues, for example price, brand, appearance or different dates, are available [10, 18, 26, 75]. Ahmed
et al. [10] also argue that the extrinsic cues, such as COOL, may be disregarded in favour of habit or preference due to
 5

Sandra Sjöö

familiarity, i.e., an intrinsic cue. This aligns with previous research that shows that extrinsic cues, such as country of
origin, play a bigger role for unfamiliar products or when food characteristics, for example taste, are difficult to evaluate
[45]. However, some research shows that Swedish consumers perceive food with labels such as ‘ecological’ or ‘Swedish’
as better tasting than conventional products, regardless of familiarity [32].
 One cue that has been particularly researched in connection to COOL is price. Despite research showing that COOL is
valued amongst consumers [32], they are not always willing to pay for it [34, 51]. According to a study from the Food
Chain Evaluation Consortium (FCEC) there is a high willingness to pay more for products with origin labelling [26].
However, it is also noted that there is a discrepancy between intention and actual behaviour when it comes to the food
actually purchased by the consumer [25, 32, 76], adding to the argument that price, as a cue, usually plays a bigger role
than country of origin. Hoffman argues that the fact that Swedish consumers are not willing to pay more for COOL
products means that the so-called Swedish model – i.e., an emphasis on high process standards, reduction of antibiotics
and strict disease control programmes – has “failed to give Swedish producers a competitive advantage” [42, p.225].
There are, however, product categories which the consumers are willing to pay more for, when the products are labelled
with their country of origin. For fresh meat products, salads, bread, frozen vegetables and vegetable oils, country of origin
has been deemed a more important attribute than for, for example, processed foods [26, 34]. The results of a comparative
study for different kinds of fish showed that Swedish consumers valued fish from the North Atlantic over fish from the
Baltic Sea, supposedly due to knowledge about environmental issues regarding certain areas [69], suggesting that origin
was a more important cue than price.

2.6 Consumer values
When choosing products, and particularly COOL products, individual values play a specifically important role in the
decision making [39]. In a study on the choice of local foods in northern Sweden, Nicolosi et al. [55] highlight that
consumers who choose cheeses produced in their surrounding area do so because they feel a strong connection to that
area and want to support the local economy. Bryła means that products and food produced in one’s own country are often
more environmentally friendly, since transportation is shorter, and chosen by consumers for this reason [18]. Lindrup et
al. highlight that food produced locally but off season is assumed to be produced with artificial heating, and is thus less
attractive to sustainability conscious consumers [49].
 Values, such as sustainability and familiarity, contribute to consumers choosing products specifically produced in their
own country [32], and, as Lindrup et al. [49] outlined in their study, online grocery shopping further enables consumers to
shop sustainably by providing access to information about the products just a few clicks away. The structure of information
is also rather unique in the online environment [49], which can aid consumers in comparing products based on specific
variables. Therefore, online grocery shopping further enhances the possibility for consumers to choose products that align
with their personal values, if that information is available to them.
 Another factor that affects consumers’ choice of COOL products is their love for their own country, their will to
strengthen and support their country, or even their unwillingness to support other countries [20]. This factor is called
ethnocentrism or cultural bias [13] and is observed to varying degrees in people [18, 45]. Juric and Worsley [45] proved
in their early work that an individual’s level of ethnocentrism plays a significant role when it comes to the their choice of
products, and argue that it should be included in all research regarding choice between domestic and foreign products. The
role of ethnocentrism is further enforced by Bryła [18], who determined that when it came to country of origin being the
most important attribute, ethnocentrism is an important aspect.
 6

COOL grocery shopping: Swedish consumers’ perception of country of origin information and labels in an online shopping environment

 Highly ethnocentric consumers pay a lot of attention to the country of origin and choose domestic over foreign products
[24, 32]. Consumers with lower levels of ethnocentrism also tend to evaluate domestic products favourably, but are more
inclined to include other cues in the evaluation, and do not reject foreign alternatives [24]. Ethnocentrism affects the
extent to which consumers pay attention to COOL in situations where little other information is available [50], preferences
for products with particular origin [75], the perceived quality [18, 24], and the “morality of purchasing foreign-made
products” [10, p.105]. Ethnocentrism is particularly evident amongst consumers who value ecological products, with over
90 percent preferring ecological pork originating from their home country over imported ecological pork [75].
 In the FCEC study, it was concluded that over 40 percent of consumers would favour domestic products if origin
information was available [26]. Thøgersen et al. [74] argue that ethnocentrism and preference for domestically produced
products is generally higher in developed countries than in developing countries, where imported products may even be
preferred. However, in situations where imported products are considered superior to domestic ones for different reasons,
such as Italian pasta or French cheese, ethnocentrism results in a more positive view of the domestic product, even though
consumers choose the imported ones [71].

2.7 Consumer demographics
Previous research has shown that consumer demographics influence COOL purchasing behaviours and preferences. Several
studies show that women choose locally produced products to a higher extent than men [18, 41, 42, 46, 69]. In their study
on the Swedish KRAV eco-label, Lannhard and Larsson [46] include women, men and ’others’, but found no significant
differences for ’other’.
 It has also been argued that consumers who prefer COOL products tend to be older [41, 67]. Shafie and Rennie [67]
highlight the positive relation between age and income, as well as education and income, and argue that older consumers
are more highly educated and therefore have higher incomes and tend to choose locally produced foods more often. Hempel
and Hamm [41] also argue that those who value and choose local food tend to have a higher educational level. Interestingly,
in their sample, Vermeir and Verbeke intentionally selected respondents with higher education based on the argument that
“they supposedly have some awareness on the concept of sustainability” and thus could provide a reliable result [76, p.178].

3 METHOD
Similar work has used both qualitative and quantitative methods, depending on what is examined. The current thesis
utilised a mainly quantitative approach in order to get an overview of the Swedish consumers’ perception of COOL online.
This has not been done on the Swedish market before, though has been conducted in similar studies in other countries
(e.g., [18]). The following section describes the method.

3.1 Survey and construction
In order to gather quantitative data for analysis, an online questionnaire was devised. As the aim of the questionnaire was
to gather quantitative data, the questions had alternative answers, specifically multiple choice, single choice and Likert
scales. Some questions allowed for free text answers, but only two of them were mandatory (discussed further below).
 As the study targeted the Swedish market and Swedish consumers, the survey was composed and distributed only in
Swedish (see appendix A).
 The survey began with an introduction, with the aim of ensuring that the respondent had previously bought groceries
online and was over the age of 18. The survey also included some definitions on labels in general, the label “Från Sverige”
and the term origin from Sweden, presented to the user in the appropriate questions (see appendix A). Questions that could
 7

Sandra Sjöö

be difficult to answer included “Do not know” or “Unsure” as options. The questions in the survey were largely based on
similar previous research (e.g., [10, 18, 24, 46, 54]), to build on existing knowledge, as well as provide an equivalent basis
for future research on the topic to continue to build upon.

3.1.1 Online shopping and stores. The first part of the survey aimed to clarify which store the respondent typically
shops with and how they perceive COOL online. For the questions regarding stores, the biggest ones in Sweden [37, 53]
were listed as options, with ’other’ as a final alternative and a free text field for entering which other. The purpose of
gathering information about the respondents’ preferred store was to allow for comparisons between how the stores were
perceived by the respondents. The questions regarding why the respondents shop online in general and with their preferred
store (Question 1 and 4, appendix A) were the only mandatory free text questions.

3.1.2 Products. Steptoe et al. [68] developed a standardised Food Choice Questionnaire (FCQ), which has been used,
referred to and adapted to by multiple studies on food and choices since (e.g., [24, 27, 52]). With the aim of making
the current study useful for other research, familiarity, adapted from the FCQ [68], was added to the options regarding
important aspects when shopping online (Question 10, appendix A).
 When the respondents were asked to rate the importance that products from a given category originated from Sweden
(Question 11, appendix A), the categories listed were ones where products can be from Sweden, i.e., coffee and tea,
to name a few, were not included. The meat products were divided into smaller categories, as it was regarded possible
that there were variations between, for example, poultry and fish. Additionally, as Sweden is one of the highest ranked
countries in dairy product consumption [33, 62], dairy products was listed as its own category, which is uncommon in
similar studies (e.g., [10, 32, 42, 45, 51]).
 Previous research has shown that price is highly relevant with regards to country of origin. Therefore, this survey
included a question regarding the respondents’ willingness to pay more for Swedish products. Importantly, the question
states “even if they are more expensive” (Question 13, appendix A), as products from Sweden do not have to cost more,
but if they do, it is relevant to know if the respondents still think they would choose products from Sweden.
 The motivation behind the choice of products from Sweden is central to the current study. To investigate the respondents’
rationale behind these choices, a question regarding how important certain aspects were to the respondents when they
chose products from Sweden. The respondents were given six statements that represented the main reasons and values
behind why consumers choose COOL products (Question 14, appendix A), gathered from the literature review (e.g.,
[18, 42, 45, 48, 74]), Svenskmärkning AB’s Swedish added values [8] as well as Jordbruksverket’s (The Swedish Board
of Agriculture) Added values in Swedish food [43].

3.1.3 The “Från Sverige” label. Section three of the questionnaire was dedicated to questions regarding the “Från
Sverige” label, and the respondents’ perceptions of the label. The familiarity with the label (Question 15, appendix A),
which the respondents rated from 1 (“Not at all”) to 7 (“Very familiar”), was of key interest. As Sweden’s only COOL, it
was deemed interesting to see the connection between the respondents’ familiarity with the label and how they valued
produce from Sweden. Additionally, section three contained questions on the respondents’ trust in the label and their
choice of products with the “Från Sverige” label above other labels (Question 17 and 18, appendix A).

3.1.4 Demographic questions. Though the survey was anonymous, demographic questions were included in order to
ensure that the study would be beneficial for future research, and to correspond to previous research. The demographic
questions were placed in the last section of the survey in order to not impact respondents’ answers to other questions,
since the demographic section included a question about diets (e.g., flexitarian and vegan). For example, if the respondent
 8

COOL grocery shopping: Swedish consumers’ perception of country of origin information and labels in an online shopping environment

knew that they answered that they are vegan, they may consider answering the question about country of origin of meat
that they simply do not eat it, however if they are not primed by this question, they may consider country of origin of meat
important if they apply it to, for instance, a partner who eats meat.
 Previous research has included the educational level of the respondents, and discussed it as both relevant and irrelevant
with regards to COOL (e.g., [10, 18, 69]), thus it was also included in the current survey. The educational levels used
in the survey were adopted from Statistiska Centralbyrån (Statistics Sweden) from their general information about the
educational levels in Sweden [64].
 Where the respondents lived was also regarded relevant, as online shopping may be more or less common and accessible
in urban areas versus the countryside. For this, respondents were asked to state where they live, with options for the
countryside as well as different size cities (Question 26, appendix A). In addition, a question about the respondents’ region
was included, with the options retrieved from Sveriges Kommuner och Regioner (The Swedish Association of Local
Authorities and Regions) [56], to provide an overview of the distribution of respondents across the country.
 As price is highly relevant when consumers consider COOL products, household income could offer another dimension
of understanding the correlation between price-sensitivity and financial circumstances. Respondents were therefore
asked to estimate their monthly household income. The options were taken from a study by Länsstyrelsen (The County
Administrative Board) developed in 2015 [40]. The options in that study were in ranges of 10 000 SEK, which were
adapted to ranges of 20 000 SEK for the current study (Question 24, appendix A), in order to simplify for the respondents.
 Special attention was given to the question regarding gender. Previous research has observed differences between men
and women and their interest in country of origin and COOL products (e.g., [18, 42, 69]), and therefore it was considered
relevant to include gender in the current study as well. However, in order to be gender inclusive, and in line with D’Ignazio
and Klein’s Data Feminism [29], ’non-binary’ and ’other’ were included as options, in addition to ’female’ and ’male’.
 For all demographic questions that gathered personal data that could be considered as sensitive, the respondents were
given the option to not answer.

3.2 Pilot study
Before the questionnaire was distributed, a small pilot study was conducted, where eight respondents filled in the
questionnaire. The respondents offered feedback on the questions, the alternative answers, which questions should or
should not be mandatory, the time spent, and any other thoughts that occured as they filled in the questionnaire. Apart
from feedback, the answers and data given through the pilot study were examined in terms of usefulness for the study, and
some formulations were altered and some questions were discarded. The answers from the pilot study were not included
in the final dataset.

3.3 Distribution and sample
The questionnaire was distributed mainly through Svenskmärkning AB’s digital communication channels Instagram
[URL] and Facebook [URL]. Additionally, it was shared through the author’s personal social media channels, as well as
those of the employees at Svenskmärkning AB.

4 RESULTS
In this section, the data collected through the survey is presented. The data has been cleansed, misspellings have been
corrected and some data, such as yes and no answers, has been encoded. In order to analyse the quantitative data, the
author has used descriptive statistics calculations, Spearman’s correlation coefficient, and Linear Regression analysis. The
 9

Sandra Sjöö

calculations were made with ready formulas in Microsoft Excel. The qualitative data generated from the free text answers
has been analysed for themes. This was done through a first rough review of the topics mentioned in the comments, which
were then refined and consolidated to an appropriate number of themes.
 After data cleansing, the survey produced 53 responses. Based on the population size of four million (40 percent of
Swedes shop online [36]), a sample of 385 respondents is required for 95 percent certainty, and 271 for 90 percent certainty.
Although an ideal sample was not achieved, the project was limited in time, and no further data could be collected. Further,
due to the small sample size, the data that was collected about the respondents’ preferred online store could not be analysed
for specific conclusions. Still, the total data suggests some interesting results, which are outlined in the sections below.

4.1 Respondent demographics
All age groups (above 18 years old) were represented amongst the respondents of the survey. The younger ages had a clear
overrepresentation, with a large part, 21 out of 53, between 18-29 years old, and 17 between 30-39 years old. A large
majority of the respondents, 42 out of 53, stated that they identified as female, 10 respondents stated that they identified as
male, and one preferred not to answer. No respondent stated that they identified as non-binary or any other.
 Among the respondents, 26 stated that they have had Post-secondary education (more than three years) and 20 stated
that they have had Post-secondary education (three years or less than three years). All respondents stated that they have had
secondary or post-secondary education of some kind. In terms of current occupation, 31 stated that they were employed,
12 that they were studying, three stated that they were unemployed, and three that they were retired. Four stated ’Other’.
The level of income amongst the respondents was rather equally distributed across the intervals. 14 out of 53 respondents,
stated that their monthly household income was between 20 000 and 39 000 SEK, though the mean monthly household
income amongst the respondents was between 40 000 and 59 000 SEK (¯
 =3,09, 
 ˜=2, =1,6).
 Out of the 21 regions in Sweden, 16 were represented among the respondents. 21 out of 53 respondents reported that
they lived in a city with over 200 000 citizens. 13 respondents stated that they lived in a city with 100 000 up till 200 000
citizens, and 10 stated that they lived in the countryside.
 The majority of the respondents, 33 out of 53, stated that they live in a household with no children (under 18 years
old), nine stated that one child was living in their household, and nine others stated that two children were living in
their household. Two respondents reported that three children lived in their household. Most respondents, 39 out of 53,
responded that they lived in a household with two adults (over 18 years old), and 14 stated that they live in a household
with only one adult (themselves).
 41 out of 53 respondents stated that they held the main responsibility for purchasing groceries in their households,
11 stated that they shared the responsibility, and one respondent stated that they had no responsibility in buying the
household groceries. 33 out of 53 respondents reported to mainly have an omnivore diet in their household, 15 that they
were flexitarian, three that they were vegan, and the final two that they were vegetarian.

4.2 Online shopping
Among the respondents, the amount of times they shopped online was evenly distributed between “Have only tried once
or twice” and “More often than once per week”. The most common alternative was “A few times per month” (¯
 =3.55,
˜=4, =1.37), with a majority of respondents shopping online “A few times per month” or more often (32 out of 53).
 
 Based on a thematic analysis of the free text answers provided by the respondents about why they chose to shop groceries
online, nine themes were identified; Logistics, Convenience, Time saving, Comfort, Pandemic, Planning, Price, Product
assortment and Sustainability. The main reasons were Logistics (20 out of 53) and Convenience (19 out of 53), for example
 10

COOL grocery shopping: Swedish consumers’ perception of country of origin information and labels in an online shopping environment

due to not having access to a car and/or not living near a larger grocery store. 15 of 53 respondents answered that Time
saving was an important reason. The Pandemic was given as a reason by eight respondents, Planning by five, Price by
another five, Product assortment by two and lastly Sustainability by one respondent. The respondents were also asked to
state why they had chosen their preferred online store. The data was cleansed and 10 themes were identified, listed in
order of identifications; Product assortment (19 out of 53), Prices (15), Habit (nine), Delivery times (seven), Membership
(seven), Service (seven), Quality (six), Delivery cost (five), Geographic availability (four), and Website (four).
 18 of the respondents stated that the stores provided no information at all about different labels and their meaning.
Interestingly, 18 were also unsure, and 17 responded that the store did provide information. 14 respondents commented on
why they thought that information was provided. The data was cleansed and six themes were identified: Information (text)
(five out of 14), General labels (four), Flags and icons (three), The "Från Sverige" label (three), Filtering options (two)
and Information on receipts (one).

 Product sustainability

 Shelf Life Country of Origin

 Price Ecological

 Brand Familiarity

 Package Design Freshness

 Use of Materials Quality

 Taste

 Fig. 1. Average value of aspects: red=total, blue=female, green=male

 The respondents were generally more positive towards the availability of information and labels (Question 7, appendix
A; “...generally see the country of origin for products?” 
 ¯=4.65, 
 ˜=5, =1.96; “...see that products are labelled with the
“Från Sverige” label?” 
 ¯=4.86, 
 ˜=7, =2.01), compared to the availability of filters for products based on the same qualities
(Question 7, appendix A; “...filter products based on country of origin?” 
 ¯=3.26, 
 ˜=1, =2.46; “...filter products labelled
with the “Från Sverige” label?” 
 ¯=3.28, 
 ˜=1, =2.51). Several respondents reported ’Do not know’ with regards to the
filtering questions (14 on each question), which are excluded from the calculations. There was small positive correlation
between the respondents’ perception of the availability of information and general interest in COOL ( =0.306), and similar
results were found between the general interest in COOL and the respondents’ perception of the availability of the “Från
 11

Sandra Sjöö

Sverige” label ( =0.403). Notably, there was a slight, albeit small, negative correlation between general interest in COOL
and the attitude towards filtering possibilities of the “Från Sverige” label ( =-0.1).
 A small positive correlation was identified between those who were generally interested in COOL and how thoroughly
they looked for an indication of origin that they relied on ( =0.217).
 Between the different aspects investigated (Ecological, Country of Origin, Product sustainability, Shelf Life, Price,
Brand, Package Design, Use of Materials, Taste, Quality, Freshness, and Familiarity), Freshness was rated as the most
important by the respondents (¯
 =6.19, 
 ˜=7, =1.47) when shopping groceries online, while Package design was rated
the least important (¯
 =2.75, 
 ˜=1, =1.63). Comparing the respondents who identified as female and male (the data was
insufficient for any other comparisons), those who identified as female tended to respond on average higher on all aspects
when shopping groceries online (Figure 1). The biggest difference in importance was for Country of Origin (female
¯=6.14, 
 ˜=7, =1.30; male 
 ¯=3.9, 
 ˜=6, =2.13). On the other hand, the only aspect which those who identify as male
rated on average higher, though only slightly, was Price (female 
 ¯=4.88, 
 ˜=6, =1.64; male 
 ¯=4.9, 
 ˜=5, =1.52). Among
those who identified as female, Freshness was rated highest (¯
 =6.41, 
 ˜=7, =1.53).
 A positive correlation was found between the respondents’ general interest in COOL and the importance of Country
of Origin when shopping online ( =0.677). General interest in COOL and Freshness had the highest correlation of the
aspects ( =0.434), except for COOL at the moment of purchase. There were also notable positive correlations between the
importance of Country of Origin when shopping online and Freshness ( =0.554), Quality ( =0.519), Product sustainability
( =0.544), Ecological ( =0.435) and Familiarity ( =0.420). Between general interest in COOL and Package Design
there was a small negative correlation ( =-0.123). Notably, Price also leaned very slightly towards a negative correlation
( =-0.102).

4.3 Interest in COOL
Overall, the respondents of the survey reported a high general interest in COOL (¯
 =5.55, 
 ˜=7, =1.73). The respondents
who identified as female reported a substantially higher interest in COOL than those who identified as male (female
¯=6.05, 
 ˜=7, =1.32; male 
 ¯=3.7,˜
 =2, =1.89). A very small negative correlation was found between the respondents’
age and general interest in COOL ( =-0.147), and a small negative correlation was also found between educational level
and general interest in COOL ( =-0.161).
 By encoding the responses with numbers one to seven with regard to the size of the city, a negative correlation was
found between the size of the place where the respondents live and their general interest in COOL ( =-0.315).
 Analyses of the data allowed for the characteristics of consumers who value COOL the highest to be identified (Table 1).

 Table 1. Characteristics of consumer valuing COOL highly

 Criteria Segments
 Gender Female
 Age 30-39
 Educational level Secondary education
 Occupation Working
 Monthly household income 40 000 – 59 999 SEK
 Place of living Countryside
 Number of adults 2
 Children in the household Yes

 12

COOL grocery shopping: Swedish consumers’ perception of country of origin information and labels in an online shopping environment

4.4 Why Swedish consumers choose COOL products
Among the different categories of groceries investigated (Bread, Fruits and Vegetables, Dairy Products, Canned and
Pickled Foods, Cereal and Muesli, Nuts, Pasta and Grains, Meat and Charcuterie, Poultry, and Fish and Seafood), country
of origin was rated as the most important for Dairy products (¯
 =6.33, 
 ˜=7, =2.25), which was also the only product
category where no respondent answered 1 on the 7 point scale. Country of origin for Nuts was rated as the least important
(¯
 =3.54, 
 ˜=4, =1.97) (Figure 2).
 Comparing between respondents, those who identified as female valued COOL, on average, higher for all product
categories than those who identified as male (Figure 2). The highest valued category amongst those who identified as
female was Poultry (¯
 =6.66, 
 ˜=7, =2.15), and the category with the lowest average value was Nuts (¯
 =3.95,˜
 =4, =2.09).
Those who identify as male value Dairy Products highest (¯
 =5.38, 
 ˜=7, =2.71), and Nuts lowest (¯
 =2.1, 
 ˜=2, =0.99).

 Dairy Products Fruits and Vegetables

 Canned and Pickled Foods Bread

 Cereal and Muesli Fish and Seafood

 Nuts Poultry

 Pasta and Grains Meat and Charcuterie

 Fig. 2. Average importance of Swedish origin: red=total, blue=female, green=male

 Out of the 53 respondents, four answered that they were willing to pay more for products from Sweden across all
product categories. 19 responded positively (“Yes” or “Sometimes”) on all product categories. There was a positive and
quite large correlation across all product categories between the importance of products originating from Sweden and the
willingness to pay more for a product originating from Sweden, should they cost more. The category with the highest
correlation was Fish and seafood ( =0.7), followed by Pasta and grains ( =0.696). Notably, Poultry had the lowest, though
still positive, correlation ( =0.393). No considerable correlation was found between the household income and willingness
to pay for the different product categories (between =-0.246 and =0.127), nor between the number of children in the
household and the willingness to pay more for groceries produced in Sweden across the investigated categories (between
 =-0.172 and =0.258).
 13

Sandra Sjöö

 17 out of the 53 respondents gave a rating of 7 (the highest possible rating on the scale) on all reasons, i.e., values, for
why they choose Swedish products; Strengthening Sweden’s self-sufficiency, Supporting the Swedish economy, Supporting
work opportunities, Food produced in Sweden is good from a sustainability point of view, Food produced in Sweden is
good from an animal welfare point of view, and Food produced in Sweden is safe for me. Among these 17 respondents,
the mean age span was 30-39, however the mode was 18-29 years old. A small negative correlation was found between
all values regarding choosing COOL products and age (between =-0.097 and =-0.329). Additionally, small negative
correlations was also found between all values and how the respondents lived (between =-0.028 and =-0.322).
 The survey showed that, among the investigated aspects of why COOL was important to consumers, sustainability
was, on average, the most important value for Swedish consumers to choose products originating from Sweden (¯
 =6.06,
˜=7, =1.59), followed by food safety (¯
 =5.94, 
 ˜=7, =1.66) and animal welfare (¯
 =5.92, 
 ˜=7, =1.81), which were
also rated highly. When comparing between the respondents’ gender identifications, those who identified as female rated
animal welfare as the most important value for choosing products from Sweden (¯
 =6.52, 
 ˜=7, =0.94), while those who
identified as male rated Sweden’s self-sufficiency as the most important (¯
 =4.5, 
 ˜=6, =2.17). Using Linear Regression
calculations, it becomes clear that the importance of the aspects all had a positive impact on the general interest in COOL.
A high regard for animal welfare had the highest impact on the respondents’ general interest in COOL (t=6.034, 

COOL grocery shopping: Swedish consumers’ perception of country of origin information and labels in an online shopping environment

 The respondents in the study rated the choice of products with the “Från Sverige” label above products with other labels
quite highly (¯
 =5.44, 
 ˜=7, =1.84). By using Linear Regression calculations, it can be noted that the respondents’ trust
had a significant impact on their choice of products with the “Från Sverige” label above other labels (t=7.629, 

Sandra Sjöö

 When considering the differences between those who identified as female and male, it should also be taken into
consideration that the respondents may, for some reason, value the scales differently (i.e., assign different grades to the
numbers on the scales), and that the results could be misguiding, in that sense, when there are differences.

5.2 Interest in COOL
In general, Swedish consumers have a high regard for the country of origin of products [32, 42], and the overall general
interest in COOL was fairly high among the respondents. However, it should also be noted that the respondents were in
large part recruited through Svenskmärkning AB’s channels, which could have an effect on the sample results, in contrast
to the general population. Additionally, the results regarding interest in, values surrounding and willingness to pay for
COOL, could be affected by the current situation in the world, as war and uncertainty has brought up threats to food
production into the media and political debate.
 Swedish legislation state that country of origin information must be available for meat, fish, honey, fruits and vegetables
[30], and, in general, country of origin is deemed most important for meat products, bread and vegetables [26, 34]. The
current study showed that meat, poultry, bread, and fruits and vegetables were rated highly in terms of importance of
being from Sweden, however dairy products had the highest rating of all. It is worth noting that within earlier studies,
dairy products are not often investigated in relation to COOL (e.g., [10, 32, 42, 45, 51]). Sweden has one of the highest
rates of dairy product consumption per capita in the world [33, 62], which could explain the category rating highly in
importance of being from within Sweden amongst the Swedish population.
 It has been discussed in research that women value country of origin more than men [18, 32, 36, 42]. The results of the
current study show that those who identified as female reported a higher general interest in COOL, as well as a higher
regard for country of origin when shopping online, than those who identified as male. It should be noted, however, that
there were a considerable number more participants in the study who identified as female. There has been very limited
discussion regarding other gender identifications and the value of country of origin (e.g., [46]) and, unfortunately, there
was insufficient data within the current study to compare beyond female and male gender identifications.
 Generally, those who regard country of origin as important tend to be older [41, 67]. However, the slight negative
correlation between the respondents’ age and general interest in COOL in the current study indicate that the younger
participants reported a higher general interest in COOL than the older individuals. Shafie and Rennie [67] also relate
age with income, and argue that older consumers have a higher income and therefore value COOL products higher. No
such correlation was found in the result of the current study, and the high representation of younger consumers within the
current study could distort the results regardless.
 Within the previous research considered, it appears to be quite uncommon to investigate how the respondents live in
relation to their interest in COOL. Bryła [18], however, highlights in their results that consumers who attach the highest
importance to country of origin live in rural areas. The current study, similarly, shows that respondents who lived in rural
areas and smaller cities tended to have a higher general interest in COOL. Additionally, slight the negative correlations
between how the respondents lived and the values suggest that those who lived in rural areas and smaller cities regarded
all values as more important than those living in bigger cities.

5.3 Online shopping and COOL
Similarly to Anckar et al. [11], the results showed that the main reasons for shopping groceries online are pricing, product
assortment, convenience and service. The main reason, however, was logistics (e.g., not having access to a car). The
results highlight that online shopping makes larger purchases possible for individuals who live further away from large
 16